Hacker News new | past | comments | ask | show | jobs | submit login

... IQ is supposed to be a normal distribution, ">= 130" === "3 sigma above the mean," i.e. top 0.1%.

(bearing in mind that the test is inescapably biased towards people with our background - US/UK/etc, certainly English speaking, probably university educated.)




To be fair, I wasn't university educated at 9, when I took my first tests, although I was born in the US and spoke English.


fair point indeed - the tests are highly weighted (and thus rather unsatisfactory - the skewing reduces the resolution of the test, if you like) for young people.


I suspect the tests I had at 9 were age-specific, but I've still kept in the same percentile.

What was interesting to me was tests a couple years ago - the questions were sometimes things like

"I'll say a number, you repeat it back to me with the digits reversed".

They kept going until you couldn't do it anymore. I think I conked out around 9 or 10 digits.

"Give me as many names as you can in 10 seconds" (something like that).

"Spot the differences between these two pictures".

I'm wondering in what ways questions like these can be (or are) culturally biased? I can think of ways myself, but curious as to what others think.




Consider applying for YC's Spring batch! Applications are open till Feb 11.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: